Coronavirus (COVID-19): Press Conference with William Hanage, 01/21/21


You’re listening to a press conference from the Harvard School of Public Health with William Hanage, associate professor of epidemiology and a faculty member in the Center for Communicable Disease Dynamics. This call was recorded at 2 p.m. Eastern Time on Thursday, January 21st.

Transcript

BILL HANAGE: Hey, everyone, apologies, I had another thing running over, I can see a lot of hands up.

MODERATOR: Yes. And I just got done doing my introductions, so I think we’re good to go. Do you have any opening remarks here today?

BILL HANAGE: No, I guess we may as well get right into it. Nice to see you.

MODERATOR: OK, great first question.

Q: Hey, thanks so much for doing this. So I wanted to ask a couple of questions about the new variants of COVID we’re seeing right now. Why do you think it’s happening now that we’re seeing this profusion of variants seemingly associated with higher transmissibility?

BILL HANAGE: That’s a very good question. I mean, to an extent, it’s not, actually now, if you think about it. We need to actually think about when they emerged, because in order to see them, they have to happen in the first place and then they have to be detected and then they have to spread widely enough that they start, you know, rearing that ugly head and people pay attention. So, for instance, B.1.1.7 was first detected in the U.K. in September but was not really noticed to be causing a problem until November. And if you do a little bit of phylogenetics, which is a sort of evolutionary tree building, and put a molecular clock on it, it seems that it actually happened or was generated some time, maybe around late summer, late August around there. I mean, there’s a bit of uncertainty about that, because some issues which are a bit too technical to go into here. But the point I’m making is that we’re becoming aware of them now, but they may well have happened at different times. Let me just say one thing before. I do want to point out something which is at the moment we’ve got three pretty much confirmed variants. I mean, there are a few others which were a little bit more unclear about. But let’s stick with those three. There have been millions and millions of infections worldwide. I think 85 million plus. And so they don’t happen every day. Seeing them now is partially because there have been so many infections and there has been more opportunity for them to occur. So that’s an important thing to remember. Now, the place I think you might have been pushing is the possibility that there could be some feature of the immune response that has been built up in the community. I don’t think we know enough to say that that’s driving any of these. However, it is certainly noticeable that at least two of them seem to have some kind of association with mutations that are expected to diminish neutralizing antibody activity.

I just saw a question, by the way, in the chat. How do you define a variant? And that’s a really good thing, actually. That’s the right way to start. Thank you for reminding me. So I’m not here talking about mutations. Mutations happen all the time. There are thousands of different mutations that have been reported that might happen against different viral backgrounds. But they don’t necessarily mean anything. Some of them may not actually help the virus at all. Some of them may make it less obvious, and it just persists a little while before that lineage goes extinct. Variants of which we’re talking about them here are not simply things with one mutation. They’re characterized by multiple ones. So all of the three that we’re talking about here have the same mutation in the spike protein, which is the protein by which the virus enters cells. N501Y is the technical term for it. And they all have that. It’s happened independently. They have independently arrived at this mutation. That mutation has also happened elsewhere in other lineages, but never taken off in the same way. Now, alongside that mutation, you can see that each of them has a bunch of others. Some of them, again, are shared between a couple of the variants. There’s one which is a small deletion elsewhere in the spike protein, which is shared between two of them. And then there’s another one elsewhere, which is putatively associated with reduced neutralizing activity. But overall, I am pretty sure these numbers are correct. I was adding them up yesterday, the B.1.1.7, which is a variant that arose in the UK, has 17 defining mutations that have changed the amino acids of the proteins made by the virus. There’s a smaller number in B.1.351, which is the South African variant of concern. I should point out, by the way, that it’s quite difficult because public health folks never want to actually say something in name of a region because of the potential stigmatization of that region. But it’s quite difficult also saying B.1.351 all the time. But the variant that has been circulating in South Africa has, I think, 9 such changes which are changing the protein sequences of the virus. And again, in P.1, which is what has been circulating in Brazil, that number is 16. So more like B.1.1.7. So the important thing to remember is it’s not just one mutation, it’s a bunch of mutations. And the significance of that is not clear. But I can quote a colleague of mine, a fellow scientist, Christian Anderson scripts, did suggest that it’s possible that N501Y is not necessarily that bad on its own, but in combination with other things can result in something which is more transmissible. Did that answer your question?

Q: Yes, I just wanted to make sure I’m understanding this right. So the common mutations as you’ve mentioned, those could be translated into greater transmissibility, but we just don’t know yet. Is that a simple way of putting it?

BILL HANAGE: I don’t think there’s any question that B.1.1.7 is more transmissible. It is actually really quite hard to come up with the way in which it’s not more transmissible than the variants that existed before. The other two it’s less clear, partially because the data is not so present on the ground. But we may well come to it. There are other reasons on top of that to be anxious about them, but B.1.1.7, that has been really pretty good epidemiological work that indicates that it is somewhat more transmissible. Exactly how much I wouldn’t get hung up on. I don’t think that that’s as important a message as the core one, the attempts that were being put in place in November in the south east of England to control the virus, which I may be wrong here, but I think the country referred to that as a second lockdown. That was not capable of controlling it. And it very rapidly became the dominant circulating lineage there at that time. So the others died out in that period. But B.1.1.7 took off.

Q: Thank you so much.

BILL HANAGE: Thank you.

MODERATOR: Next question.

Q: Hi, thanks. I wanted to ask about President Biden has a goal on vaccinations of 100 million shots in the first hundred days. And I just wonder, I mean, do you think that that’s high enough? I’ve heard some estimates that, you know, if we’re doing that great a million shots per day, I guess it would take some time into next year before we reached whatever the herd immunity threshold for the US is. So, do you think we should be, I guess, shooting higher than that or is that even possible to shoot higher than that? Or what do you make of that goal?

BILL HANAGE: So I’m going to start by commenting that I’m not a sort of supply chain vaccine rollout expert. But I can talk to the questions of how many people need to be vaccinated. By the way, now we’ve got vaccines, we can move away from talking about herd immunity and start talking about the critical vaccination threshold, which is a much nicer way to be talking about it. Now, one thing about what I was just saying, which is important, that if you have more transmissible virus, what it means is that you need to vaccinate more people before you control it. The simplest relationship is really quite easy to describe. In reality, there may be things which are more complicated because of contact patterns. But regardless, you need to vaccinate more people. And I think that it’s reasonable to think that the proportion of the population that would need to be immune is probably closer to 80 if we are talking about something like B.1.1.7. Now, one hundred million is not nothing, but it would be less than a third of the US population. Importantly, though, and this may be a part of the message, although obviously I don’t have any insight into that. It would include a relatively large proportion of the older population who are most vulnerable. And I think a lot of places are having a goal like vaccinating those people for very good reasons because they are the most vulnerable to infection by far. But also, I want to point out, as I think you realize from your question, it’s only partway towards the sort of levels of immunity that we would need to actually control the virus. And you could still get considerable amounts of transmission in younger age groups. And younger age groups, while they’re much less likely to suffer the most severe consequences in terms of dying, are quite capable of having both long COVID relatively large fraction of them and or being hospitalized. And so in terms of the long term burden on health care, we need to be thinking carefully about that.

Q: Got it. Thanks. And without B.1.1.7, you know, with the current strain, I guess, what percentage do you think? Would be less than 80 percent?

BILL HANAGE: Yeah, it would be closer to 60. I mean, I don’t want to get too specific with these numbers because the fact is, they’re based on extremely naive models and they’re like for illustration purposes. The true number in both cases is probably a little bit less, but probably not that much less. The reason for that, just out of sheer nerd, this is the only opportunity I get to nerd out in epidemiology and help is to do with the contact networks that people make and the fact that they are heterogeneous, meaning that some people make a lot of contacts and some people make far fewer. And vaccinating the people who make the more contacts is obviously going to be more effective than vaccinating the people who make fewer.

Q: Thanks.

MODERATOR: Next question.

Q: Thank you, Dr. Hanage. In addition to the three variants of concern that have been outlined by WHO and recognized by the CDC, there have been a slew of preprints, not a slew, three published within the last week or so, talking about variants perhaps around the Columbus, Ohio area and one from Southern Illinois University, talking about a variant that probably emerged in the US in Houston in May and makes up about a third, I think of the cases here. And then there are some unpublished research we’re waiting on out of California discussing a variant that they think they have out there that might be associated with the surge of cases there. And I’m wondering if the gene changes and those variants look as concerning as the variants of concern that have been recognized by WHO. Are they as worrisome or do we even know? Is it too early to know that?

BILL HANAGE: So that’s a really interesting question. It’s interesting, does three count as a slew? I think I would caution again, as you know, I always caution when thinking about preprints. As I said previously, a lot of the mutations that we see in the three variants have been observed before in isolation or in combination with a couple of other mutations. And they have been going around for a while and have never taken off to the extent that the ones with concern now are taking off. So I think that that’s really important when it comes to interpreting these things. Among those specific cases, to my eye, the cases from Ohio look like they are mutations of concern that have happened on the background of circulating viruses. And I’m not particularly anxious about those given the data I see there at the moment. Another point worth making is that in a lot of the places in the US, we’re still getting a tiny fraction of genome sequences. So while in the south east of England, in Kent in November, the rise of B.1.1.7 was first identified as a cluster of more than 100 anomalously closely related cases with no known sort of thing linking them together. So that’s how that was first identified and that’s when they were sequencing about five percent of all the infections there. In contrast, the stuff from Columbus, they’ve managed to do a few dozen from the last month or so. And I’m really pleased, I think it’s great that they’re doing sequencing and looking for it at all. But I just don’t think that’s a large enough sample to be making any strong conclusions, especially when you if there is an outbreak which is associated with, say, a long-term care facility, then that in and of itself can drive a transient increase in the number of closely related things. And if that happens to have one of these mutations that people might look at, then you might feel like you were worried for a bit. But you need to spend more time looking at it. In terms of southern Illinois, a variant rising in May, which has been going around since then, is not really a matter of as much concern as something which is increasing more rapidly recently. And as I read that preprint, there’s a proposal there which I don’t think I’m quite sold on yet. But it’s interesting that that variant might have been associated with less severe disease. I think that that has to be disentangled from the nature of the age groups that are getting infected, better treatments and a whole bunch of other stuff like that. So, again, I put that aside. The variant from, I suppose, a variant from California, and again, the numbers are too small really to be jumping to any conclusions. I’ve also not been able to figure out whether or not the mutations are happening alongside others in the way of the three variants that WHO has been focusing on so far. So California, I think, watch that space. But the others, I think, are not variants of the kind that we’ve been talking about more recently.

Q: Thanks, that’s really helpful, Nicole, can I ask one quick follow up question?

MODERATOR: Sure, go ahead.

Q: I’ve seen a lot of calls among academics in the past few days, that now is the time we need to be thinking about updating vaccines to adjust them for the presence of at least the variance of concern. Do you agree with that? And how long would that take?

BILL HANAGE: Oh, it wouldn’t necessarily take long. The reason I say yes is because given the fact that we have discovered that SARS-CoV-2 is capable of producing these sorts of things, it’s reasonable to start preparing our vaccine approaches, to be able to deal with it. And that is not, by the way, I’m sure we’ll get to this. But that is not me pronouncing at all on whether or not I think the currently available vaccines will be effective against a variant. I just think that ensuring that we have the capability to produce a vaccine which is efficacious against these or any other variants that might pick up in the future, because the fact that these happened three times already means we can expect them to continue happening pretty much. Now, one of the good things about the mRNA vaccines from Pfizer and Moderna is that it would not be terribly hard to modify the technology in order to be able to produce a vaccine against a different spike protein. So that is something which is an optimistic thing. Of course, on the other hand, those vaccines are among the most expensive and most difficult to roll out logistically and so on and so forth. I’m less sure about the other important vaccine at the moment, AstraZeneca. But I do think that it’s important for us to be in this position where we are ready to think about doing this, and of course, there really is a very good example of this already existing, which is every year flu vaccines get updated based on whatever the properties of circulating lineages, circulating strains are. And so I think that given that we are able to do that already, we’ve kind of intellectually set up to do that for flu. It’s a completely reasonable thing to think that we should be preparing to consider it for SARS-CoV-2.

Q: Thank you.

MODERATOR: Next question.

Q: Hey there, I’m just wondering if you can kind of give a bit of the lay of the land of how the outbreak looks in the country as a whole to you right now? I mean, it seems like some of the numbers show maybe we’re coming down from a post-holiday peak. Obviously, maybe the variants will affect this. Could you just explain to people kind of where you see us right now in late January?

BILL HANAGE: Yeah, that’s actually a really great question. I mean, the first thing to say, as you know, as we’ve been saying all along, is that within the country as a whole, there are all sorts of local stories. But I think the overall sort of take-home message is that the following the holidays, we did see a burst of transmission, which was higher in some places than others, but that we have now got to a stage, it’s not that it’s exactly coming down very quickly anywhere, but it is sort of stabilizing in various places. I mean, within that, there are other very particular sort of stories that are locally important. And indeed, the locally important ones often are a result of local transmission in sometimes restricted to particular groups and health care, which is one of the reasons why California has been struggling in the way that it has is because of basically access to the beds and that health care, which is capable of dealing with the people who are most critically ill. I mean, the other thing which we saw, and this is in some ways a reflection, what happened previously. But you’ve probably seen these things, and I think it’s worth looking at again. In some states, particularly in the northeast, I haven’t checked Massachusetts recently, but based on the trends from our going, I think Massachusetts is among them, so is New Jersey, so is New York. More than one in every 500 people have died from SARS-CoV-2 already. Now, that is largely as a result of what happened last year in the spring. The vast majority of those happened then, but they have been ticking up. Interestingly, the Dakotas, having had virtually very little pandemic activity until October, very rapidly shot up the league table over the course of November and December. But recently, other states have been sort of adding to the totals that they picked up over the earlier part of the pandemic, which is not surprising because we’re in winter and transmission is expected to be somewhat higher in the winter and people are expected to make more contacts indoors. So the fact that it’s stabilizing is only part of the story. The fact is it stabilizing at high levels, and this is something else I want to just touch on a little bit, because I think this is very important. When we come back to variants, one approach to pandemic response which I think has been followed by a large number of places in the United States and a few in Europe as well, has been to try and control the level of transmission at a level which is does not require very, very severe restrictions. But also, doesn’t go exponential and overwhelm health care. Now, if you set that, I mean, I’m not saying I agree with that, but it is one kind of approach to pandemic management. The problem which comes with that is if you are running that really high close to health care capacity or the point when it becomes seriously compromised, then any change in the underlying epidemiology of the virus, whether it’s because of the people moving indoors or the arrival of a new variant can have pretty grave consequences, because it takes you from that point where you’re capable of dealing it to the point where you’re not.

MODERATOR: You all set?

Q: Yes, thank you very much.

MODERATOR: OK, great. Next question.

Q: Hi and thank you. Is what you just described, what you think is happening in California, you know, we’ve had this exponential surge in cases and deaths since mid-November. And I think people are really struggling to understand because we have a lot of restrictions, a lot of mask wearing. I just wonder if you think any of the new variant or variants, they’re talking here about L452R, Cal22c, could they play a role in what we’ve seen and if so, like, how do we find out?

BILL HANAGE: Can I first say how much I like the image, what I’m seeing on my screen at the moment. That’s awesome. I don’t think that we can come to that conclusion about California at the moment. I think you need more data to really figure that out. I will note that among those states which have reported the most, B.1.17, California is up there. I think California and Florida, are the ones right next to Bangkok and the on the top of the screen. California and Florida are among those which have reported the most of B.1.1.7. So if it’s not an issue now, it may be in the future. The other thing to remember, though, about California is that when you’re talking about restrictions and the like, those restrictions will have been necessary at a different stage in the outbreak in California as a result of that health care capacity, and I mean, that’s why it’s such a difficult tightrope to walk. I mean, we are able to put these things kind of dial up and down the dimmer switch. But how effective they are depending very much on the extent of compliance. It depends very much on the extent of the types of contacts that people are making. The communication I think needs to focus on things which are risky activities and what actually constitutes risk and whether or not the risk to yourself or risk to others. So I do sympathize with you, but I don’t think there’s evidence to say that California has been struck by a variant that is yet undetected. As I say, I’m a bit unconvinced by some of these reports that are coming out. I mean, if I see more data, I might change my opinion. You know, I should point that out. And I should also point out that I think that if there was anything circulating in the United States and California, in fact, in particular, because it has some very good genomic epidemiologists there, at the kind of level that’s happening in the U.K., we might know about it or we probably know about it, at least in California. I do think that the fact that this country does that kind of epidemiology and genetic epidemiology on only one tenth the scale that the United Kingdom does, it means that there could be some nasty surprises circulating among us that we don’t know about yet, and we should be prepared to see them if we start looking harder. Does that make sense?

Q: Yeah, thank you very much.

MODERATOR: Next question.

Q: Thank you very much for doing this. On the point of finding new variants or doing surveillance. I’m just wondering what impression you have, if any, about whether the new administration or any of the agencies or indeed academic labs and so on are expanding this kind of, you know, fingerprinting of the viruses that we’re so inadequate at or have been when we have all of the apparatus to do it based on fluids.

BILL HANAGE: I could do a whole call on this alone, so I’m going to try and summarize it as quickly as I can. The new administration is certainly invested in improving this part of the country’s response. And that’s good. That’s really good. The challenge with that comes from things which are very deep seeded and embedded within the health care system in the United States, and I can illustrate it like this. If you want to do sequencing, if you want to get a genome, you have to take a sample and you have to isolate RNA from it. That’s the same thing as you’d have to do for a PCR test. So you’d think that testing or tests would be a great source of stuff for genome sequencing. Well, because testing is often done by companies rather than by groups of academic or governmental labs, they have no incentive to even retain enough RNA sequencing, and that means that it’s hard to coordinate the sort of approach which has been seen elsewhere. There have been some interesting examples of collaborations between some of those companies and CDC, which is very helpful. Sort of interest, and this is one of the reasons why there’s a whole like kind of on background thing here. If you look at the CDC’s flu set up, which is in many ways extraordinarily impressive, it’s also evident that it has the challenge in that it has to work with the states so you can only get things from states that they are willing to sort of share with you or send you. And if a state is not willing to do something, then you can’t do anything about it. And as an illustration of that, North Dakota, despite having one of the worst per capita death tolls in the entire pandemic, as far as I’m aware, I think the first genome sequence of the virus from there came out less than two weeks ago. And this is an illustration of the challenge which CDC has had for very many years. It’s an extraordinary organization, but there’s only so much it can do.

Q: But I mean, I guess is it a question of will on the part of the states or resources that they lack to do this? And is that something that I mean, is this another question of like underfunded skeleton, public health, dealing with too many things at once?

BILL HANAGE: To an extent. And it’s a sense of prioritization. I mean, if you ask me, I think it would be great to have a national genomic epidemiology service. I would like that to be sentinel labs coordinating in each state. I think that there are a lot of things that could be done like this, but it does depend somewhat on the states willingness to cooperate in the powers of federal government to set things up in the individual states and the individual states priorities because the reasonable amount of public power is devolved to the states. When it comes to the question of will, I do think that there’s a big issue with that and I have I am hopeful that we may be seeing a pronounced shift in tha.t and I mean, the way I think about it is that I’m not American by my accent, but back in the early days of cell phones, it was always really confusing when Europeans came to the United States and they just sort of looked around and nobody had a cell phone. And then, of course, the iPhone came along, and the United States became completely addicted to cell phones. In somewhat of the same way, I’m somewhat hopeful that just maybe having seen this can be what can be done with this elsewhere and how important it is to the detection of early threats and characterizing them, the United States will now go the same way with this type of epidemiology.

Q: Thank you.

MODERATOR: Next question. She wasn’t sure if you’d answered this already, so I’m just going to ask in case I missed something you said that you said. You say that mutations we’ve seen converging on the same point in the viral genome means we’ll see more. What’s this virus trying to do? How can we thwart its evolution in these directions, not just masks and social distancing, but vaccinations and strategies? I think you touched on that a little bit. And do you see this coronavirus being endemic like flu?

BILL HANAGE: There’s a lot in there. Let me start, it’s not so much the convergent evolution that makes me say we will see more. It’s the fact that we have seen three of them and we will see others, convergent or not. Now, what the virus is trying to do, because those of us who work on the evolution of infectious disease, evolution, epidemiology, infectious disease often falls into the trap of saying somebody is trying to do something but is not selected, having done something. So any virus which transmits to more people will be expected to leave more descendants. There are a few wrinkles in that which are interesting, and which are clear for B.1.1.7, again, not clear for the others. You will also see selection on a virus which transmits more quickly if you imagine you are infected with the virus that infected 10 people over the course of one year before eventually you cleared the infection and it was competing with another variant of the same virus, which only infected people for a week and infected two people in that time, while the second one would outcompete the former pretty quickly because it would be spreading more quickly. And we that it doesn’t look like that’s happening with B.1.1.7, incidentally, it looks like B.1.1.7 is actually just more transmissible. It’s not that it’s got a shorter generation time.

Then there’s a lot of questions about whether or not viruses are adapting to be more virulent. That’s very unclear. The there’s lots of talk about this. And my friend Marc Lipsitch once made a comment that was extremely amusing and I completely agree with it, which is that the ratio of theory to data in that field is very poor indeed. There’s a lot more theory than there’s data. And so I think that we shouldn’t get into that. I think that we should just focus on something like the best way to stop it, which was the next part of your question is to stop it. And we stop it using the combination of distancing, other non-pharmaceutical interventions, vaccinations. If you can get a decent testing strategy to keep things low, then that helps, and then vaccines. That’s the way to do it. And finally, you ask whether I thought it was going to become like the flu. And I think that with that in mind, you do have to remember we have only ever eradicated one virus, smallpox. And so I think it’s reasonable to assume that we will not eradicate SARS-CoV-2. What our goal is going to be is to control it to the extent that we don’t get damaging surges of infections, which leads to unnecessary illness and loss of life, and in due course for it to become a virus, which is among those that cause that afflict humans, because there are lots of viruses, sort of like humans that just, you know, not pandemic viruses. And when it’s no longer a pandemic virus, that would be a very, very, very different situation.

MODERATOR: Thank you. I think she’s going to be all set. Next question.

Q: Yeah, hi, Bill, can you hear me?

BILL HANAGE: Yeah.

Q: Thanks for doing this. I wanted to ask about kind of the generalizability of the functional change for some of these various mutations, like do we expect that any lineage that independently co-evolves mutations at these locations in this spike protein will also be more transmissible and eventually become dominant in the populations where they arise? Or is that also dependent on local conditions in terms of preexisting immunity?

BILL HANAGE: That’s a really smart thing. It provides an opportunity as well right at the end to say something else important, which is that all the things we see depend on local conditions. It’s very hard to disentangle what a virus has been doing from what it has been given the opportunity to do, if you know what I mean. I think the answer to your question is we don’t know really, but what it indicates is the importance of this concept from genetics, which is called epistasis. In epistasis, the effect of the mutation depends on other things within the genome. So you often see this in terms of antibiotic resistance, where you have a mutation which confers resistance, but which also comes at some cost to the organism. And so it might spread for a while among patients taking a lot of antibiotics, but then eventually dies out. Unless another mutation is present, a so-called compensatory mutation which ameliorates that cost, lessens that cost. And in that case, the impact of the resistance mutation is different depending on what else is happening in the genome. And it may be that something like that is happening here, where I mean, there are some proposals in the literature which go into protein folding, which are frankly beyond my area of expertise, but which suggests it may be the N501Y, which we know increases the affinity of binding to the ACE2 receptor which the virus uses for the cells, it may be that on its own it causes something else in terms of the protein structure that makes the resulting variant less than the resulting virus less fit overall, but that is compensated by some of these other changes and exactly what that is, I suggest you go and talk to somebody who deals with protein folding and or who is more of a virologist than an epidemiologist. But I think it’s quite telling that the variance that we have seen so far have these quite striking properties of having not only N501Y, but also others.

Q: Thank you, can I just ask one to follow up question? It seems like the N501Y confers this increased transmissibility fitness advantage where some of the other ones we’ve seen confer this antibody escape advantage. Is there anything we can see in epidemiological data that would give us some insight into what evolutionary forces the virus is responding to, whether it’s, you know, prevalence of immunity from natural infections versus just being able to get more inside a newer host?

BILL HANAGE: Well, sort of. I don’t think we know enough about what’s going on in South Africa yet, but it is extremely noticeable that P1 arose in Manaus, in the Brazilian Amazon. Now, Manaus had a really, really tough time around, like around May, June last year with very with a severe surge and a very rapid number of deaths. And then the virus suddenly fell away in June. And a lot of people wondered if this was a consequence of herd immunity, that what had actually happened was the virus had infected so many people that there just were not enough people left who were not immune for it to be able to transmit effectively. And lots of people talked about this and it was discussed in terms of people suggesting herd immunity strategies and so on. And just last week, I think maybe before there was a paper published in Science which suggested that as much as 75 percent of the population of Manaus could have been exposed. And the basis for that was a combination of studies and levels among blood donors and also with some modeling to sort of account for the fact that titers wane a little bit over time, so people might have been infected in the past and you wouldn’t pick them up if you were testing them in September or October. So that may be a little bit high, but it does suggest a very large amount of immunity in terms of notes, but now this is currently suffering another crisis of health care and is running out of oxygen and has been appealing for help from neighboring Venezuela in order to deal with the numbers of patients it has. And the fact that this is happening in a place which has been previously exposed to such large amounts of immune transmission and which we infer therefore has substantial immunity, is extremely worrying, very worrying indeed. And I’m aware as yet of no evidence to suggest that P1 is more likely to be, in fact, it’s quite difficult to know how you tell that because of the fact that people say that to definitively prove a reinfection, you have to have definitive proof that a person was infected in the past and that you need to have sequenced the original virus and then you want to show that they’re not that different things. So I’m not aware of any such evidence. However, it’s one of the things that you want to be thinking about very, very carefully as we are going forward. I’ll just say, in passing about South Africa, I’m not aware of any evidence that the South African variant is certainly causing infections. However, I do think that there’s reason to be quite concerned, given the interactions with convalescents here. But this is something which I think we want to get into some good epidemiological data. Finally, B.1.1.7, as I may have said already, I don’t think we have evidence that B.1.1.7 is sidestepping immunity. In parts of the UK, there isn’t much other virus left. But if you look at people, the rates of infection among people with B.1.1.7 versus other stuff, they’re pretty similar, so I don’t think there’s much evidence that variant is sidestepping. I’m going to say I would be much more concerned about the other two.

Q: Thank you.

MODERATOR: I had a quick question. She’s wanted clarification, we were speaking about updating the vaccines. Do you mean right now to adapt to variants or as a booster shot if people get a booster shot annually or both?

BILL HANAGE: Huh, interesting. I think that right now we need to be on the front foot with this, as opposed to an annual vaccination that is going to depend on things that we’re going to learn about the way in which the virus behaves when a large portion of the population has vaccine induced immunity. But at the moment, I’m suggesting being on the front foot to handle any variants, rear their ugly heads. You know, if I’m still doing this in October, it’s probably better to ask me the other question closer to then.

MODERATOR: I don’t know what to say about October. All right, next question.

Q: Hey, Dr. Hanage, thanks for making the time, as always. As you noted, we kind of share that dubious distinction with California here on the B.1.1.7 variant. And I did want to ask you about that. It’s hard to get an idea of how many sequences are, the CDC isn’t releasing that when it’s putting out its numbers of cases identified in various states. And also, the state hasn’t been very forthcoming about its health department efforts here. Locally, we have a research university. The University of Miami is partnering with Jackson Health, which is our public hospital, and they’ve developed some custom PCR assays that they’re going to use to try to identify the variants in hospital patients. So that will give us maybe some idea if and when the variants get here. But my question is, you know, here in Florida, our hospitalizations have started to kind of level off, if things take off again here and if cases and hospitalizations start to rise quickly, and that coincides with the discovery of some of these variants here in Miami. I mean, is it safe to assume that the variant is what’s causing things to go back up at that point? Or how do you parse those factors out when you’re looking at that?

BILL HANAGE: Yeah, it’s a good question. And the first thing to note is I think that this is some context which you may be aware of and what you’re talking about. Miami is not actually that overwhelmingly different from what has been happening elsewhere. A nice thing about some of these variants, and I say nice, I mean a convenient thing about some of these variants is that. One of the mutations they contain means that they give a very characteristic sort of test result when you’re testing for the virus, using certain tests, not all tests, but some tests. And that’s the kind of shift that you’re talking about there. People are saying that they’re going to be running these tests and then they’re going to look at the results and anything which looks funny, they will rapidly sign for sequencing.

Q: This is actually different. I did ask about that, but the research lab here has designed custom PCR assays for every known variant. So the P1, the B.1.1.7, and then they’re also going to be sequencing whole virus. So that’s going to be coming later. But for right now, they just want to hit a bunch of clinical samples and look for the variant, because I don’t think they feel like the state is doing enough to do that. So that’s kind of the efforts that they’re taking here.

BILL HANAGE: I think that is a good idea. The only question I have is in order to detect something when it’s rare, you often need to go over a whole bunch of stuff which is important. And I don’t know enough about the work. But I would also comment, and they may well have done this, it’s the combination, again, of things which is there, so I don’t know how for what you’re talking about there, are they doing multiple PCR things? And so you’d like the readout from the multiple PCRs?

Q: In other words, they’ll get a positive sample, and they will then run a PCR that looks for the P1 or a PCR that looks for the B.1.1.7 on those hospital patients who have already tested positive to try and identify if the variant is there in the hospital, basically.

BILL HANAGE: I think that’s a good idea. The comment I was making, by the way, was because of the fact that the environment characterized by multiple different mutations, you may have to do like multiple PCRs in order to identify the multiple mutations there. So you get your test to get a positive result and then you do a combination of other PCR things. Right? One thing that could be done, which I mean, it depends on the sort of sequencing technology that you’re using. There are sequencing technologies that you can use where you can amplify up the spike and then put it onto a nanopore, which is a teeny tiny little sequencer, about the same size as that. That’s not a sequence of events, but it’s literally about the size of that and it gives you very rapid results. So this does sound like something which is worthwhile. The only caution I put on there is that, like I said, if something is present at one percent, or if something is rare, you have to sequence quite a lot in order to be able to get an accurate impression of exactly how common it is. And getting it when it is rare is a lot better than getting it when it’s common, you know.

Q: You know, if we are to see kind of a resurgence here, like I said in the last week, maybe two weeks, we’ve seen hospitalizations level off. How do you suggest kind of reading any future spikes, given the fact that we know B.1.1.7 is here? I mean, should it be the first question we have in mind when we look at cases spiking again after a relative plateau?

BILL HANAGE: Yeah, I think that if you see marked increases from a plateau of that kind, I suggest that you should seriously consider whether or not B.1.1.7 is responsible. This is certainly the situation that happened in the southeast of the United Kingdom, which has been very, very hard, indeed. And it’s again, it’s that combination of trying to have a plateau of comparatively high hospitalizations, which then suddenly gets hit with a wave. Sorry, I don’t like the word wave. We’re doing with the surge, it’s not able to contain as a result of the fact that the underlying epidemiology has changed. I will point out, however, that there are a lot of I mean, it’s very difficult to separate these things out from the local conditions. Florida is notable in multiple ways, part of it to do with the you know, with specific demographics among groups who are particularly at risk. But also, obviously, because of the fact that it makes it easier for you guys to meet outside than it is for me to go and meet people outside. So it may be that the number of at risk contacts is somewhat lower. But if you were to see a sudden change, I think that it’s reasonable to suggest that you would want to be looking very carefully to see if B.1.1.7 was there. And obviously, you then have to think about taking action against it in a way that get the governor of Florida has historically been quite reluctant to do.

Q: Yeah, you took the words right out of my mouth.

BILL HANAGE: Yeah, I had a feeling I might have.

Q: Thank you, Dr. Hanage, I appreciate it.

MODERATOR: Next question.

Q: Dr. Hanage, thanks for doing this. I came a little bit late, so I don’t know if you answered this in an earlier question, but I was wondering if you had any practical advice for people who are maybe more concerned about the variants that are more infectious now. I’ve seen some writing that suggest that people double mask. Is that something that you would suggest people do or are there other things, other measures people can take in their lives?

BILL HANAGE: I think that the important place to start with this is remembering that the variants are not some kind of super magic virus. They’re not so different that we shouldn’t be able to reckon that we already know quite a lot about how to handle them while going about our lives. I think that it’s accurate to say that increased vigilance, improved mask use, is certainly something that will be helpful. The specific example you raised of double mask use; I think that I’d defer to my colleague Joe Allen to look into to explain that more thoroughly than I could. But there are indications that some combinations of masks are superior to one on their own. But getting into all the very many different combinations of masks that are possible is completely beyond the time that we have here, or indeed my expertise. But the thing is, if you were already doing things that put you at risk of being infected by the circulating lineages of SARS-CoV-2, your risk of being infected by B.1.1.7 is higher, because of the fact that you are having those risky exposures and the way to handle that, if you’re anxious, is to reduce the number of risky exposures, and you can’t do that by a number of things which have been part of public health advice for months now. And I think that it’s worth reiterating that and worth making clear that people do have it within their power to reduce their risk and to continue reducing that risk.

Q: Thank you. I do have one final question, if you don’t mind. Governor Baker, today, our governor here in Massachusetts, of course, said that he is loosening some of the restrictions that he put in place earlier since I think that we’re seeing a bit of a plateau in cases here in the state. There are still plenty of sort of mitigation measures, measures that he’s taking place. But it’s a very slight easing. Do you think that sends the wrong message, given our concerns about new variants and the fact that this pandemic is not going away any time soon?

BILL HANAGE: That’s a very good question. So this is against a context where Massachusetts has gone through, roughly speaking, over this winter, three stages initially that were around three thousand five hundred cases a day. And then Thanksgiving happened. And in the weeks after Thanksgiving, that led to a stable five thousand cases day. And then the holidays happened at the end of December and beyond that again and is now seeming to be declining a little bit. And I think that what that indicates is that absent these other events, these gatherings, the reproductive number of Massachusetts is around about one, which means you’re not getting exponential growth. With that in mind, I can understand the motivation to be very cautiously stepping back. And one of the things which is interesting about Governor Baker’s approach is that it has been much more cautious than an awful lot of places. And thus far it has managed to escape the sort of exponential surge. The thing which makes me anxious about it is the fact that relaxation is expected to give the opportunity for B.1.1.7, which in Massachusetts, it’s rare, but I mean, as the governor said himself, we should assume that it’s here, the opportunity to spread more. And because of the fact that hospitalizations, as you correctly note, they have actually sort of tailed off and they’re starting to drop a little bit, which is very good. It runs into the problem that we’ve been talking about previously on this call. But it runs into the problem that if you’re already operating at a very, very high baseline of pandemic activity, relatively small changes, can have a disproportionate impact, and I think that it’s very important in any relaxation to emphasize that it is entirely possible that those restrictions can be put back in place if cases start to increase and the risk to public health returns. I mean, the risk to public health is extremely present anyway. You know, I’ve actually, as I said a few times on here, Massachusetts has seen one in five hundred residents die from this already. So it’s something which we should take extremely seriously, given that it’s like January and we’ve got a fair amount of cold weather to get through. In fact, it’s Massachusetts, we’ve got a load of cold weather to get through. It never gets warm here until like June.

Q: Yeah, thank you.

MODERATOR: Next question.

Q: Thank you again, like everyone said, for making the time today. So there’s kind of a split screen happening now with these newer, faster spreading variants emerging and this kind of painfully slow vaccination process to watch. So, I mean, how do you think these interact and what are the next six months to look like? Do we start to see partial effects of partial herd immunity along the way, or does it kind of turn on like a light switch?

BILL HANAGE: It won’t turn on like a light switch. And I can guarantee you that what we’re seeing now is, and I want to point out that what I’m about to say to you now is obviously a general sort of thing for expectations, it’s not intended to be a very precise or exact statement of what if it’s not a forecast, OK, but it is something you can see for spectacular expectations as more people are vaccinated, we can expect to see that we will get more bang for our buck when it comes to the other interventions that we are having to live with, because some people are not going to be being infected because they’re vaccinated rather than because they’re not making contacts with other people who are infected, which is great. But that’ll be gradual. The exact time at which things become quite normal, it’s probably much further off than most people appreciate it and there are several reasons for that. One of them is that the variants, they present a very real risk of a sting in the tail. And they will require, as I said, they at the very least, the presence of B.1.1.7 will require more people to be vaccinated in order to achieve protection. We can then expect there to be a shift, and in fact, I predict that the story is going to shift from being the sort of, oh, just let everybody get infected and somehow will the elderly and the vulnerable off to being once a certain fraction of the elderly and vulnerable are vaccinated, then everybody else should go back to normal. And that’s not really going to be something which is advisable, because for a start, the vaccines are effective. But we don’t know how long that effectiveness lasts. We don’t know. And even worth bearing in mind, even 95 percent efficacy still means that one in 20 people would become infected. And given the consequences of infection in the oldest age groups, that would still be something which would be risky.

But I’m more concerned about the potential transmission and burden on health care and consequent burden of long comfort of lots of transmission among younger people. So what I think is going to happen is that over the next few months, enough people are going to be vaccinated, that it’s going to be noticeable, that we don’t get these large fluctuations in transmission that come with more opportunities for transmission, and then that’s also going to coincide with the spring because as we saw from the spring last year, that reduces it doesn’t eliminate the ability for infection and transmission, but it does seem to reduce it. Possibly people are more able to make more contacts outside. That seems very plausible to me. And what will then happen is that cases will continue to drop down as people are vaccinated. I think my estimation for a point at which we’re going to kind of look at each other and say, while all of this is so much better than it was, is probably around the middle of this year, but it’ll be a gradual process. And I also want to hold out the possibility that we might need to, I mean, this depends upon a number of factors all working well, that includes they’re not being really worrying things, for instance, about variance or about vaccine rollout not going really badly. It has been going really badly, but I think there’s reason to think that it’s going to improve and any of the other sort of unknowns. So if you ask me what it’s going to look like, it’s going to look gradual. I think that actually, most of my colleagues thought that January was going to be the worst month and I was among them. I think that it’s plausible to suggest that givenB.1.1.7, we should be prepared for February to be really bad as well, and possibly depending on vaccine rollout, even March and April as well. So I think that we have to recognize that exactly. In fact, in some ways I feel like you asked the question in a way that was better than I have managed to answer it, which is that you have this balance between vaccine rollout and this race against time, against the new variant B.1.1.7 in particular, because we know it’s present. And what happens over the next three months is going to depend on who wins that race.

Q: Thanks. And to follow on that, you kind of touched on this, but can you address, I know I don’t want to drag you into the politics of this, but if we have immunized the most vulnerable to death in the population, seems like the political calculation of the open up, virtual lock down argument kind of shifts. So would you expect that political pressure to boil over in some way?

BILL HANAGE: Yes, I think I would. It’s a really good question. The reasons to be cautious about it are those which I outlined because younger people can get seriously ill. And if a large number of people in the under 50 age group get sick quickly, then we expect that to be a severe burden on health care. In fact, another comment I can make on that is Rochelle Walensky, who’s hopefully going to be confirmed as the next head director of CDC, and others wrote a paper in JAMA Network a few weeks ago, I think, which was estimating that in some parts of the country, which has seen a lot of disease, COVID-19 is now the leading cause of death for the 25- to 49-year-old age group because people in that age group really do not usually die. So you have to remember that those risks continue. And among those things is, however, if I can provide what I would like to see if I were in a political position to be able to achieve these things a carefully sort of cautious step by step return, including things like masks, because, as I said, as more people become immune, the more bang for your buck. And so it’s possible that you might be able to get more thank you were. Nicole has put it in the chat. And so you might be able to tolerate not full reopening, but something which is closer to a more open economy, because that is where we want to get back to, it’s just we don’t want to get back to it in such a way that it increases the risk to public health.

Q: That was great, thank you.

MODERATOR: A quick question. I’ve heard that people with weakened immune systems who can’t quickly fight off the virus might be breeding grounds for variants. Do you agree?

BILL HANAGE: This is a really tough question. And one of the reasons why it’s tough is that I don’t think anybody would use words like breeding grounds. I think that it is, and I know this because I was involved in one of these studies, we have looked at some long-term infections of people who are immunocompromised in some way or other. And we have noticed changes in their genome sequences that look similar to some of those that have happened in the variants. Indeed, they are the same as some of those of the variants. Patients that we examined, we studied here in Boston, had the N501Y mutation which arose during the course of their infection. Now, the important thing there, and the other thing to note is that if you look at these variants, that phenomenon where they look like they’ve had more mutations, and you would expect, given the time when they were isolated, does suggest evolution in a somewhat different selective setting, rather like you might expect if somebody was infected for a long period of time, chronically and chronically exposed to elements of the immune system, but not all of it. However, it’s also true that evolution to do well in a host, to make more copies of virus in a host, is a different environment from trying to get between hosts. And it’s kind of like how can I put it is a good metaphor. It’s kind of like a redecorating your home rather than finding another home. And what that means is that you don’t necessarily expect things like that to become more transmissible. So it is possible that these chronically infected, immunocompromised hosts, are one of the places where variants the origins of arrogance, but we don’t know. It’s also worth noting that people rarely, very rarely, people who are not immunocompromised can be infected for a long period of time. And so I have to say, I think the answer is that firstly, you should never use the word breeding ground because of the stigma that can come with it. And secondly, chronic infections of one kind or another are a prime candidate for where things of this nature could emerge from. But the exact nature of that or exact nature of those infections remain to be discovered because we have never actually seen one of these things in the act of emerging, we’ve only seen it once it’s come out and reached a sufficiently high level in the population.

MODERATOR: A quick follow up on that one. Does that mean that selective pressures coming from monoclonal antibody treatments, or convalescent plasma treatment as well?

BILL HANAGE: Yeah, it does. There is a study of the so-called London patients, which, again, you shouldn’t call people London patients like that. Hence why I say so-called, in which it was a patient with published treated with convalescents here. And among the mutations which that patient developed, that deletion, that 69, 70 and the spike protein, which is also found to B.1.1.7, and is the reason why it produces the characteristic test results. So once again, it’s in somewhat early days and I think we need to urgently study the cases of chronic long-term infection in order to better understand what is happening there, because it certainly is true that the clock rates, which is the rate with which new mutations substitutions accumulate, is faster within an infection than between an infection.

MODERATOR: OK, thank you. Next question.

Q: Hi. Thank you so much for doing this and thank you for taking my question. So the variant that has been reported in California, specifically in Los Angeles, is and I think you referred to this earlier, but again, how much credence do you give to the idea that a single variant that is only now being detected in California might be responsible for its surge?

BILL HANAGE: Not much, but I will make some comments and I’ll make actually comes back to the origins of B.1.17. The first the most people knew about B.1.1.7 was when the British health minister stood up in the House of Commons and started talking about this variant. And a number of people just kind of turned around and said he’s just trying to blame that because they’ve done such an appalling sort of job of pandemic response. And my own personal reaction was that it was possible for more than one thing to be happening at the same time. So in the case of California, I think time remain, it remains to be seen, but I do think that you can never entirely separate out the presence of a circulating variant from exactly what’s going on the ground. And so, I mean, we’ll know more as we study it more. But I think that it’s pretty clear I don’t think that there are a lot of items are circulating variants in California. I don’t think there have been in the past. So, you know, California has been having problems going back a while, and so even though it’s been having problems going back a while, and hypothetically they get worse now because of a variant that still doesn’t explain the problems that they have been going back a while. If you think about what’s going around L.A. Does that make sense?

Q: Yes, yes. Thank you.

BILL HANAGE: You’re welcome. And then you have a question.

Q: Thank you so much for this opportunity and thank you for acknowledging my hand there. Talking with some of the scientists here in Texas, the state and the country as a whole, they say that we’re not doing enough genomic sequencing and that what is being done isn’t readily shared among everyone in the scientific community. So what does this mean when it comes to looking at these types of variants of concern in the United States?

BILL HANAGE: It’s a good question. The United States doesn’t do anywhere near enough genomic sequencing. I mean, even in places that do a lot of it, it’s not the kind that you would exactly. What you really want is a large proportion of random cases to be sequenced as it is. People like me cobble together partnerships with places and try and bag them for RNA in order to get sequences or indeed there’s targeted studies which is devised for something else. I mean, in the US, the ability to do genome sequencing is very much linked to whether or not there are samples collected and then there is an academic center nearby with the wherewithal to be able to do the sequencing. And then once that happens, you have to decide what to do with it. And that means making it publicly available. And there have been, over the course of the pandemic, varying extents to which people make their data publicly available. In some cases, there are decent reasons for that because there is sort of aggregators which do a pretty good job of taking data from publicly available data sets and then making them available so that people can sort of explore them and people then want to hold onto it. However, I think that in a pandemic, we probably ought to just reward everybody who’s doing any sequencing and making the data available as quickly as possible. I think the incentives are misaligned. But the important thing going forward is that we should be doing more sequencing. We are probably going to get better at it. But it’s something which, like we were saying earlier, takes political will. And I think a large part of what happens with these ongoing sequencing efforts is that the data that comes out of them should be made publicly available in the early stage. The one caveat I have on that is that interpreting it sometimes takes a little bit of expertise, just as, for instance, if the sequence generated isn’t very good quality, it can sometimes look like it’s got a lot of mutations. And as we’ve been saying, people look at a lot of mutations and they sort of a spider sense starts tingling. So if you were to look at that, you know, that’s an example of the sort of why it still does need expert interpretation. But on the other hand, that’s not hard to achieve. I mean, we could set this up pretty quickly if there was the will.

Q: And on that will, just a quick follow up. Would it be the willingness of the United States or would it take something like the World Health Organization? I know that they’re pushing for a sharing of information, but also the same sort of naming of the variants as well across the across the world.

BILL HANAGE: Oh, yeah. The nomenclature thing is sort of amusing. So in case we’re talking here, I think it’s the United States which needs to do it. I think it’s very important for the United States to do it. I also think it’s important for the world to do it. And as you may be aware, the World Health Organization published a really, really, really good sort of primer on how to set up a genomic surveillance system, which I highly recommend to anybody who really wants to know more about this. It was extremely good. And of course, the data do need to be shared not only within the United States, but around the world. So I think that on the first level, you know, right now the US is one of the countries, certainly among the wealthy, highly resourced nations that has the worst genomic surveillance. And that should change. And then there’s a longer battle, which is about figuring out how to make us better globally in terms of improving our ability to identify threats and respond to them. And I want my final comment on this that and respond to the matters, because you have noticed over the last hour or so, I’ve talked a lot more about B.1.1.7 than I have around, B1351, or P1, even. One of the reasons for that is that there’s been good epidemiological work done on it. And that has been possible because the UK has more resources, among other things. And that is something which is the next part of the piece of this. We have to come up with ways of where we’re not only doing the sequencing, but we are getting the other what we call metadata from it as well, like the age of the patients, the characteristics of their disease and so on and so forth. Those were all very important parts.

And the one thing you said at the end that’s kind of funny, there is confusion arising over nomenclature because a paper that was published nine months ago suggested a fractional nomenclature, which is what the B.1.1.7 refers to. Separately, the folks who run the next strain website, which is a great introduction instantly to the genetics of virus of viral pathogens, have proposed a separate naming system based on clades which have names like 20A and they’ve named the variants SN501YV1, which is, it’s a separate nomenclature that’s the same thing. These things which next named are the same as the things which are named B.1.1.7 and stuff like that. It’s just a little confusing when they are not aligned with each other, in particular to the public. And also, I have to say, as a person who often talks to the press about these things, it’s also a real pain to be saying B.1.1.7 all the time. That would be a lot easier if they could be given something which was a little bit easier to trip off the tongue. But I will be honest, in a pandemic that is not the not the highest and most important of my concerns.

Q: Thank you so much, I really appreciate.

BILL HANAGE: Thank you.

MODERATOR: Dr. Hanage, do you have any final thoughts for us?

BILL HANAGE: Other than that, I really hope that in the nicest way in the world, I really hope to not be talking to you. But unfortunately, I’m not sure that’s going to be the case.

This concludes the January 21st press conference.

View more press conferences